PSI - Issue 19

6

Marc J.W. Kanters et al. / Procedia Structural Integrity 19 (2019) 698–710 Marc K nters et l. / Structural Integrity Procedi 00 (2019) 000– 00

703

Figure 5: Stress strain curve for the plate samples taken at 0°, 45° and 90° (solid lines), compared with the model fits (dashed lines; stiffness only)

4.1.2. Localization factor During experimental material characterization, in general, force and displacement signals are obtained during the measurement and subsequently converted to stress and strain. Depending on the geometry, however, stress and strain can localize in specific regions. Consequently, the local values will deviate from the nominal. In FE analyses the local values are computed and this can lead to inconsistencies between simulation and experiment. For example, when using a stress- or strain-based failure criterion combined with a material model based on nominal input data, it will trigger premature failure. To overcome this issue, the localization factor, � , should be used during material calibration to transform the macroscopic values to local values. The stress-based localization factor is defined as the ratio of the local stress at the hot spot and the nominal stress. In this study, the stress localization factors have been computed for each fatigue load case. The factor depends on specimen geometry and microstructure. Figure 6 illustrates the stress localization factor workflow, from localization to scaling of the SN-curve, to calibration of the final fatigue model.

Figure: 6. Illustration of the workflow to correct for the localization factor